It is now widely accepted that the TGF-B signaling pathways can both suppress tumor formation and promote tumor progression and that both effects are mediated largely through tumor cell autonomous TGF-B signaling. Data generated during this grant period with conditional knockout of TGF-B signaling in epithelia support the hypothesis that epithelial cell autonomous TGF-15 signaling is tumor suppressive and demonstrate that metastases can not only occur but are enhanced with knockout of the type II TGF-B receptor (Tgfbr2) in mammary carcinoma cells. However, systemic inhibition of TGF-IS signaling markedly suppressed pulmonary metastases in the MMTV-c-neu/DNIIR mice. This indicates that the effect of systemic inhibitors of TGF-IS in suppressing metastasis is largely on host cells. These results have modified our hypotheses concerning the mechanism of TGF-B promotion of tumor progression. We now propose that tumor cell autonomous signaling by TGF-li can suppress rather than enhancing metastases, and that TGF-B signaling in host cells is a significant component of both the tumor suppressive and the tumor promotion effects of TGF-fi in vivo. We will test these hypotheses through the following specific aims by determining the changes in the carcinoma cells and their microenvironment associated with knockout of the type II TGF-li receptor gene, Tgfbr2, in tumor cells that lead to increased metastases. Specific Aim 1. Determine the effects of systemic inhibition of TGF-IS signaling on mammary tumor formation and metastases from MMTV- c-neu and MMTV-PyVmT-induced mammary tumors in the context Tgfbr2 knockout in mammary epithelial cells effected by both MMTV-Cre and WAP-Cre. Specific Aim 2. Determine the influence of timing of loss of TGF-B signaling during mammary tumor formation and progression on metastatic spread using inducible MMTV-Cre. Specific Aim 3. Determine mechanisms for enhanced metastatic capability with loss of tumor cell TGF-li signaling by examining both tumor cells and their microenvironment.